Skip to content

Latest commit

 

History

History
89 lines (54 loc) · 2.48 KB

README.md

File metadata and controls

89 lines (54 loc) · 2.48 KB
Raw

EavesDroid: Eavesdropping User Behaviors via OS Side Channels on Smartphones

This repo contains the source code and dataset of our paper "EavesDroid: Eavesdropping User Behaviors via OS Side Channels on Smartphones" published in IEEE Internet of Things Journal (IoT-J).

EavesDroid

0x01 Getting Started

Prerequisites:

  • Android Studio
  • Android NDK
  • Python 3.11
    • numpy
    • matplotlib
    • keras
    • django
    • cydtw

0x02 Repo Structure

1. Android App

  • app/ contains the source code of our Android app.

  • Please rename it to Sampler and import it into Android Studio to build the app.

  • Note: val api = "http://192.168.1.2:8000" in MainActivity.kt should be changed to the proper server address.

2. Server

  • backend/ contains the source code of our Django server.

  • Please run python manage.py runserver to start the server.

3. Data Collection

  • collection/ contains the source code of emulated data collection tool.

  • Please run python sample.py/sample2.py to collect data from your own devices.

  • Note: parameters should be changed to the proper values.

4. Data Classification

  • classification/ contains the source code of user behavior classification tool.

  • all_model.py contains the baseline models used in our paper: 1D-CNN, LSTM, GRU.

  • dtw_model.py contains the DTW-KNN algorithm used in our paper.

  • classify.py is the main script to infer user behaviors with our CNN-GRU model.

4. Dataset

  • dataset/ contains the dataset used in our paper.

  • See Releases for downloading the original dataset dataset.tar.xz .

  • Data in this directory can be used to train classification models and reproduce our results.

5. Figures

  • figures/ contains the figure generation scripts used in our paper.

  • Files in this directory can be used to reproduce the figures in our paper.

0x03 Copyright and License

This project is licensed under the terms of the MIT License.

0x04 Contact and Citation

If you have any questions, please contact me through GitHub Issues or email: wangquancheng@whu.edu.cn.

If our work is useful for your research, please consider citing our paper:

@ARTICLE{wang2024eavesdroid,
  author={Wang, Quancheng and Tang, Ming and Fu, Jianming},
  journal={IEEE Internet of Things Journal (IoT-J)},
  title={EavesDroid: Eavesdropping User Behaviors via OS Side Channels on Smartphones},
  year={2024},
  volume={11},
  number={3},
  pages={3979-3993},
  doi={10.1109/JIOT.2023.3298992}
}